Modulation of μ-opioid receptor desensitization in peripheral sensory neurons by phosphoinositide 3-kinase γ

Neuroscience 169 (2010) 449 – 454

MODULATION OF ␮-OPIOID RECEPTOR DESENSITIZATION IN PERIPHERAL SENSORY NEURONS BY PHOSPHOINOSITIDE 3-KINASE ␥ C. KÖNIG,a O. GAVRILOVA-RUCH,a G. SEGOND VON BANCHET,b R. BAUER,a M. GRÜN,a E. HIRSCH,c I. RUBIO,a S. SCHULZ,d S. H. HEINEMANN,e H. G. SCHAIBLEb AND R. WETZKERa*

the critical targets of opioids acting on opioid receptors, which are abundantly expressed in the cell body and terminals of DRG neurons. Particularly the ␮-opioid receptor (MOR) was found to induce heat analgesia (Scherrer et al., 2009). The interaction of opioids with MOR is accompanied by increasing desensitization of the G protein-coupled receptor (GPCR), which has been linked to the induction of tolerance, a phenomenon of extraordinary importance in medicine (Bailey et al., 2009). Mechanistically, opioid receptor desensitization involves receptor phosphorylation by different cytosolic protein kinases and interaction of the phosphorylated receptor with arrestin eventually leading to receptor internalization and recycling (Schulz et al., 2004; Koch et al., 2005; Koch and Höllt, 2008). Previous studies have reported an involvement of phosphoinositide 3-kinase (PI3K) dependent signaling pathways in the sensitization of primary nociceptive neurons (Bonnington and McNaughton, 2003; Zhuang et al., 2004). For example, pharmacological investigations using broad specificity PI3K inhibitors have proposed a role for PI3K activity in MOR desensitization in sensory neurons (Tan et al., 2003), but the identity of the PI3K species involved is unknown. All four members of the class 1 PI3K family produce the second messenger phosphatidylinositol-3,4,5-trisphosphate (PIP3), which controls a multitude of cellular functions (Hawkins et al., 2006). The only class IB member PI3K␥ was characterized as a major mediator of GPCR agonists. As holds true for other PI3K enzymes, PI3K␥ exhibits both a lipid kinase and protein kinase activity (Bondeva et al., 1998). As yet PI3K␥ has been documented in immune cells (Hirsch et al., 2000; Fruman and Bismuth, 2009) and cells of the cardiovascular system (Patrucco et al., 2004). In addition, a number of studies have reported expression of PI3K␥ in neuronal tissue (Narita et al., 2002; Cunha et al., 2010). Given the proposed involvement of PI3Ks in MOR desensitization and the well-established function of PI3K␥ as a signaling mediator downstream of GPCR in leukocytes (Fruman and Bismuth, 2009), we tested the possibility that PI3K␥ might act as a mediator of desensitization processes induced by prolonged opioid treatment. Herein we report reduced pain relieving effects of morphine and strongly decreased tolerance development after systemic morphine application in mice deficient of PI3K␥. Taken together with accompanying electrophysiological recordings in DRG, our findings disclose a novel, neuronal function of PI3K␥ as a specific mediator of opioid tolerance development and MOR desensitization in the peripheral nervous system.

a Department of Molecular Cell Biology, Center for Molecular Biomedicine, Jena University Hospital, Jena, Germany b

Institute of Physiology I, Jena University Hospital, Jena, Germany

c

Molecular Biotechnology Center, University of Torino, Torino, Italy

d Institute of Pharmacology and Toxicology, Jena University Hospital, Jena, Germany e Department of Biophysics, Center for Molecular Biomedicine, Jena University Hospital, Jena, Germany

Abstract—G protein-coupled opioid receptors undergo desensitization after prolonged agonist exposure. Recent in vitro studies of ␮-opioid receptor (MOR) signaling revealed an involvement of phosphoinositide 3-kinases (PI3K) in agonist-induced MOR desensitization. Here we document a specific role of the G protein-coupled class IB isoform PI3K␥ in MOR desensitization in mice and isolated sensory neurons. The tail-withdrawal nociception assay evidenced a compromised morphine-induced tolerance of PI3K␥-deficient mice compared to wild-type animals. Consistent with a role of PI3K␥ in MOR signaling, PI3K␥ was expressed in a subgroup of small-diameter dorsal root ganglia (DRG) along with MOR and the transient receptor potential vanilloid type 1 (TRPV1) receptor. In isolated DRG acute stimulation of MOR blocked voltage-gated calcium currents (VGCC) in both wild-type and PI3K␥-deficient DRG neurons. By contrast, following longterm opioid administration the attenuating effect of MOR was strongly compromised in wild-type DRG but not in PI3K␥deficient DRG. Our results uncover PI3K␥ as an essential modulator of long-term MOR desensitization and tolerance development induced by chronic opioid treatment in sensory neurons. © 2010 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: MOR, PI3K␥, DRG, VGCC, desensitization, opioid tolerance.

The powerful pain relieving effects of opioids were originally assigned to the activation of opioid receptors in the central nervous system but recent investigations also demonstrate an important role of peripheral mechanisms of opioid analgesia (Stein and Lang, 2009). Peripheral sensory neurons in the dorsal root ganglia (DRG) are among *Corresponding author. Tel: ⫹49-3641-9395600; fax: ⫹49-3641-9395602. E-mail address: [email protected] (R. Wetzker). Abbreviations: DAMGO, D-Ala2,N-Me-Phe4,Gly5-ol-enkephalin; DRG, dorsal root ganglion; GPCR, G protein-coupled receptor; MOR, ␮-opioid receptor; PI3K, phosphoinositide 3-kinase, class 1; TRPV1, transient receptor potential vanilloid type 1; VGCC, voltage-gated calcium channel.

0306-4522/10 $ - see front matter © 2010 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2010.04.068

449

450

C. König et al. / Neuroscience 169 (2010) 449 – 454

EXPERIMENTAL PROCEDURES Nociceptive assay, dose–response data and tolerance induction The in vivo experimental procedures were performed on wild-type and PI3K␥⫺/⫺ mice (10 male and female each, 11–13 weeks of age, 24⫾4 g). Animals were housed four to six to a cage with same sex littermates and maintained at 12-h dark/light cycle in a temperature-controlled environment with unrestricted food and water. The wild-type and PI3K␥⫺/⫺ mice were derived by 10 generations of successive backcrosses of heterozygous male knockout mice from chimeric C57BL6/129Sv PI3K␥⫺/⫺ mice (Hirsch et al., 2000) with C57BL/6 females (Jackson Laboratories, USA). Experiments were approved by the committee of the Thuringian State Government on Animal Research. All testing was conducted near mid-photophase (9 AM to 4 PM) to reduce circadian effects on nociceptive and analgesic sensitivity (Kavaliers and Hirst, 1983). Following a 30 min habituation to the testing room, mice were assessed for baseline nociceptive sensitivity on the 45 °C tail-withdrawal test. In this assay of acute, thermal nociception, the mouse is gently restrained and the distal half of the tail is immersed in water maintained at 45.0⫾0.2 °C by an immersion circulator pump. Latency to reflexive withdrawal of the tail was measured twice to the nearest 0.1 s, with each determination separated by a minimum of 20 s. The two determinations were averaged. The tail-withdrawal test was chosen because of its stability even after repeated exposures to noxious water temperatures (D’Amour and Smith, 1941). A cut-off latency of 45 s was employed to prevent tissue damage. Immediately following baseline latency assessment, subjects were injected with an initial dose of morphine (1.0 mg/kg), followed in succession with increasing doses (2.0, 3.6, 6.5, and 11.7 mg/ kg). For tolerance development, morphine was applied at increasing dosages (20, 40 mg/kg) at two following days with three injections each day. At day 4, repeated tail-withdrawal test was performed with the same escalating morphine administrations. Percentage of the maximum possible effect (% MPE) was calculated by the formula: 100%⫻[(agonist response time⫺basal response time)/(cut off time⫺basal response time)]⫽% MPE.

Immunohistochemistry Sections (5 ␮m) of paraformaldehyde-fixed and paraffin-embedded DRG tissues from adult rats (Wistar), wild-type and PI3K␥⫺/⫺ mice were labeled with anti-PI3K␥ monoclonal antibody (Leopoldt et al., 1998) and mouse anti-IgG2a (Sigma) as isotype negative control. Colocalization of PI3K␥ was analyzed on adjacent slices with anti-TRPV1 polyclonal antibody (diluted 1:100, raised in rabbit against a 15 AA peptide of rat transient receptor potential vanilloid type 1 (TRPV1) receptor of the cytoplasmatic C-terminus of the receptor, Alpha Diagnostics, Germany) and anti-␮-opioid receptor polyclonal antibody (diluted 1:5, raised in rabbits). Dewaxed sections were rinsed with ddH2O and transferred to an autoclavable cuvette with 10 mM citrate buffer. Heating for 15 min at 120 °C was used for antigen retrieval followed by PBS wash and blocking step with 2% serum and 0.3% Triton X-100 in PBS for 30 min at room temperature. Primary antibodies incubated over night at 4 °C and were detected using biotinylated secondary antibodies (diluted 1:100, Dako, Denmark) and the ABC-Vectastain-KS Kit (Vector Laboratories) following manufacturers instructions. Visualization followed with application of the peroxidase substrate Jenchrom (MobiTech, Germany). In the case of PI3K␥ colocalization with MOR, we used secondary antibodies labeled with Cy2 and Cy3 (diluted 1:200, Dianova, Germany). Control experiments were carried out with omission of all primary antibodies.

Analysis of immunohistochemical data The sections were examined with a light microscope (Axioplan 2, Zeiss, Germany) coupled to a CCD video camera and an image analyzing system (KS 300, Zeiss, Germany). The mean area, diameter, and mean grey value were determined for each neuronal soma. To take into account differences in the basal grey values of individual sections, a relative grey value of each neuron was calculated by dividing the mean grey value of the neuron by the grey value of the cover slip background. For an unbiased discrimination of cells with or without positive labeling with antibodies, all neurons were considered as positive if they showed a relative grey value above that of neurons from the control incubations, which were not treated with the primary antibodies. Proportions of labeled neurons are expressed as the mean⫾SD. In double labeling experiments sections were additionally analyzed with a standard fluorescence filter. Neurons were classified as unlabeled, single, and double labeled.

PCR Total-RNA extracts of murine 48 h primary DRG culture, WEHI (mouse myelomonocytic leukemia) and MEF cells (mouse embryonic fibroblasts) were isolated with RNeasy-Kit (Qiagen, Germany) and cDNA was produced using TaqMan Kit (Applied Biosystems) following manual instructions. Polymerase chain reaction for PI3K␥ (forward: 5=-GGAGAACTATGAACAACCGG, reverse: 5=-ATCTCACTTCGCAGGAAC, product 2100 bp) and ␤-actin (forward: 5=-GAGGTATCCTGACCCTGAAG, reverse: 5=CAGAGGCATACAGGGACAG, product 250 bp) ran 95 °C 30 s, 59 °C 30 s, and 72 °C 2 min for 40 cycles.

Primary DRG culture DRG were dissected from whole spinal cord of wild-type and PI3K␥⫺/⫺ mice (C57/B6J) and collected in DMEM/F12 (Invitrogen). Digestion followed with collagenase (type II, 400 U/ml, PAA, Austria) for 45 min and Trypsin/EDTA (0.05%, PAA, Austria) for 10 min at 37 °C. DRG were suspended in DMEM/F12 and cells were dispersed by mechanical trituration with a fire-polished Pasteur pipette. The suspension was then plated on Poly-L-Lysine (50 ␮g/ml, Sigma) coated coverslips in DRG medium (DMEM/ F12⫹10% heat-inactivated FCS supplemented with 1 mM L-Glutamine (Invitrogen), 1% PenStrep (PAA, Austria) and 100 ng/ml nerve growth factor (ProSpec, Israel). Measurements were performed within 48 h of culture.

Transfection of primary DRG neurons Freshly isolated DRG cell suspensions (ca. 105 cells per transfection) were electroporated using Amaxa Rat Nucleofection Kit and Amaxa Nucleofector I (Lonza, Germany) following manual instructions. Plasmid DNA of pEGFP-N1 encoding wild-type PI3K␥ (1 ␮g, PIK3CG, GeneID: 5294) were combined with plasmid DNA encoding for the transmembrane glycoprotein CD8 (0.3 ␮g). After electroporation cells were seeded dropwise onto Poly-L-Lysine coated coverslips. Positively transfected DRG neurons were preselected by CD8 specific Dynabeads (Deutsche Dynal, Germany) and thereafter identified via green fluorescence.

Electrophysiology Whole-cell voltage-clamp recordings were performed with an EPC-9 amplifier (HEKA Elektronik, Germany). Pulse protocol generation and data acquisition was controlled with PatchMaster software (HEKA Elektronik, Germany). Series resistance errors were compensated to 70%. Correction for leak currents and capacitive transients was executed using a P/6 method. Data were low-pass filtered at 5 kHz. For whole-cell voltage-gated calcium current measurements external solution was composed of (mM): 140

C. König et al. / Neuroscience 169 (2010) 449 – 454

451

TEA-Cl, 10 CaCl2, 2 glucose, 10 HEPES, (pH 7.4 with CsOH). The patch electrode with 1–2 M⍀ of KIMAX-51 glass (Kimble glass, USA) was filled with (mM): 105 CsCl, 2.5 MgCl2, 5 glucose, 10 EGTA, 10 HEPES, 2 Mg-ATP, 0.5 Na2-GTP (pH 7.4 with CsOH). Whole-cell calcium currents were evoked by 100 –150 ms depolarization steps from ⫺80 to ⫹10 mV. D-Ala2,N-Me-Phe4,Gly5-olenkephalin (DAMGO, Sigma) and morphine (Mundi-Pharma, Germany) were stored as 1 mM stock solutions in 10 mM HEPES at ⫺20 °C. Before application with a glass pipette, ␮-opioid-receptor agonists were diluted in external solution. For ␮-opioid receptor desensitization, DRG cells were preincubated for 6 h with either 2 ␮M morphine or 2 ␮M DAMGO in DMEM/F12 medium and shortly washed with external solution before measurements. AS605240 (1 ␮M, Alexis, Germany) or bis-indolyl-maleinimide (1–5 ␮M, Sigma) were added to medium 30 min before recording.

Statistical analysis Data are reported as means⫾SEM. In in vivo experiments, twoway repeated measures ANOVA was used to examine the main effects of genetics and dosages, and their interaction, on tailwithdrawal baseline and dose-dependent latencies. Post hoc comparisons were made with the Holm-Sidak test. Comparisons between groups were made with unpaired t-tests using Bonferroni correction for multiple uses, when appropriate. Differences were considered significant for P⬍0.05.

RESULTS Antinociceptive behavior and opioid tolerance development of PI3K␥ⴚ/ⴚ mice To elucidate a possible incidence of PI3K␥ in the nociceptive and antinociceptive behavior we performed a comparative tail-withdrawal test of the PI3K␥⫺/⫺ strain and wild-type animals (Fig. 1). No differences were observed in the initial response to heat stimulation. Mice showed a basal reaction latency of 6.64⫾1.11 s for wild-type and 7.58⫾0.76 s for the PI3K␥⫺/⫺ strain. Systemic application of escalating morphine doses at day 1 increased the response latency both in wildtype and PI3K␥-deficient animals. However, PI3K␥⫺/⫺ mice showed a reduced response in the range of 5 mg/kg morphine (P⬍0.05, lower panel). At day 4 after chronic morphine treatment, we noticed profound differences in the magnitude of the antinociceptive effects. As expected, wild-type animals showed distinct tolerance development in comparison to day 1. In contrast, the attenuation of morphine-induced analgesia was significantly less pronounced in PI3K␥⫺/⫺ mice (P⬍0.05, middle and upper panel). These data unveil PI3K␥ as a mediator of morphine-induced tolerance development in mice. Expression of PI3K␥ in dorsal root ganglia The analgesic effects of morphine are mediated by opioid receptors, which are found throughout the nervous system, in the brain, in the spinal cord and also in primary sensory neurons. We asked, therefore, in which neuronal tissue PI3K␥ is colocalized with ␮-opioid receptors (MOR). Whereas tissue samples from the central nervous system did not display any detectable expression of the catalytic subunit of PI3K␥ (data not shown), we found high immunoreactivity (IR) for PI3K␥ in small-diameter DRG of rats and mice, which hold for about 20% of peripheral sensory

Fig. 1. Differential morphine tolerance development of WT and PI3K␥⫺/⫺ mice. Morphine analgesic tolerance was assayed in WT and PI3K␥⫺/⫺ (n⫽20) mice. Morphine dose-response curves were obtained by cumulative morphine administration at day 1 (bottom) and repeated escalating morphine application at day 4 (middle). The maximum possible effect (% MPE) and its difference (top) are presented as means⫾SEM (* P⬍0.05).

neurons (Fig. 2A). This expression pattern confirms recent data (Cunha et al., 2010). Importantly, and in extension of these data, histoimmunochemical analysis of adjacent DRG tissue slides revealed that PI3K␥ was expressed in a subgroup of heat-sensitive and nociceptive neurons colocalized with MOR and the thermo-sensitive cation channel TRPV1 (Fig. 2B). Thus, from the analyzed rat neurons 31.5% showed a TRPV1-like IR and 22.4⫾4.9 % of the neurons were labeled with the anti-MOR antibody which is in accordance with previous data (Endres-Becker et al., 2007). Out of 800 analyzed cells 91.3%⫾4.8 showed coexpression of PI3K␥ and TRPV1. About 67% of PI3K␥and MOR positive cells showed a colocalization of both antigens. MOR expression in almost all peptidergic substance P-positive murine DRGs has been demonstrated in previous studies (Scherrer et al., 2009). The average diameter of PI3K␥ positive cells was 16.3⫾6.9 ␮m (in rats) and 12.3⫾7.1 ␮m (in mice) suggesting that these were small neurons predominantly forming C-fibers. PI3K␥ expression in primary dorsal root ganglia was confirmed by RT-PCR experiments in which WEHI (mouse myelomonocytic leukemia) and MEF served as positive/negative control, respectively (Fig. 2C). In sum, these data demonstrate coexpression of PI3K␥ and MOR in a subtype of peripheral sensory neurons of DRG.

452

C. König et al. / Neuroscience 169 (2010) 449 – 454

Fig. 2. Coexpression of PI3K␥ and MOR in dorsal root ganglia. Immunohistochemical stainings (A, B) were performed on sections from paraformaldehyde-fixed and paraffin-embedded DRG from rat, as well as wild-type and PI3K␥⫺/⫺ mice with anti-PI3K␥ monoclonal antibody, mouse IgG2a-isotype control, anti-TRPV1 and anti-MOR polyclonal antibody. (A) DRGs from rat and wild-type mouse showing immunoreactivity (IR) for PI3K␥ in morphological small-diameter neurons compared to IgG2a control staining and PI3K␥⫺/⫺ DRG. (B) Colocalization of PI3K␥- and TRPV1-like IR as well as PI3K␥- and ␮-opioid receptor-like IR on consecutive slices of rat DRG providing evidence of expression in nociceptive neurons (scale bars: 25 ␮m). (C) PCR for PI3K␥ and ␤-actin on cDNA from murine 48-h primary DRG culture was compared to cDNA of WEHI (mouse myelomonocytic leukemia) and MEF cells.

PI3K␥-dependent MOR-induced desensitization of voltage-gated Ca2ⴙ channels To address the impact of PI3K␥ on MOR-dependent signaling on a cellular level, we isolated primary murine DRG cells and performed whole-cell patch-clamp measurements of opioid-sensitive voltage-gated Ca2⫹ channels (VGCC). Single small-diameter neurons with a capacity of about 20 pF were selected for the assays. The current densities of wild-type and PI3K␥⫺/⫺ cells were indistinguishable. Blocking the activated voltage-gated Ca2⫹ currents by acute application of morphine or DAMGO (2 ␮M) inhibited the Ca2⫹ currents in almost all cells of both wild-type and PI3K␥⫺/⫺ DRG cells in a similar manner (Fig. 3A for DAMGO, normalized values for both agonists in Fig. 3C, E). Long-term MOR desensitization was induced by preincubation with morphine or DAMGO for 6 h. Preincubation with DAMGO and subsequent acute application of 2 ␮M DAMGO resulted in a significantly decreased inhibitory effect of the MOR agonist on Ca2⫹ currents, consistent with a functional desensitization of the MOR signaling pathway (Fig. 3B vs. Fig. 3A, all measured cells included into statistical analysis). In contrast, applying the same treatment regime in PI3K␥⫺/⫺ neurons evidenced an almost unperturbed ability of DAMGO to reduce Ca2⫹ currents, indicating that 6 h DAMGO preincubation did not result in a desensitization of the pathway in PI3K␥⫺/⫺ cells (Fig. 3B, normalized values in Fig. 3C). Similar results were obtained with morphine as the agonist (Fig. 3E). The virtual absence of the opioid-induced desensitization in PI3K␥⫺/⫺ cells suggests an essential function of the signaling protein in MOR-induced desensitization. To address whether or not these effects were mediated by the enzymatic activity of PI3K␥ we subjected DRG cells from a transgenic mouse strain expressing a kinase-dead PI3K␥ variant (termed kinase dead, KD; Patrucco et al., 2004) to the same protocol. As shown in Fig. 3C (black bars), KD mutant cells behaved as PI3K␥⫺/⫺ cells in that chronic

opioid exposure did not hamper VGCC inhibition by acute DAMGO application, strongly indicating that the enzymatic activity of PI3K␥ was critically involved in MOR desensitization. In order to ascertain that the observed effects were a direct and bona fide consequence of the absence of PI3K␥, we performed reconstitution experiments. Transient PI3K␥ expression in small-diameter PI3K␥⫺/⫺ DRG cells restored desensitization of MOR signaling leading to a hyporesponsiveness to acute DAMGO administration comparable to wild-type cells (Fig. 3D). Consistent with the genetic data, pharmacological inhibition of PI3K␥ with the specific inhibitor AS605240 compromised MOR desensitization in wild-type DRG cells, as evidenced by an almost unaffected ability of DAMGO to exert an acute VGCC current block (Fig. 3F). We also addressed the involvement of classical PKC, since PKC reportedly regulates MOR internalization (Kelly et al., 2008), and PKC␣ was recently been identified as a downstream mediator of PI3K␥ (Lehmann et al., 2009). As shown in Fig. 3F the PKC inhibitor bis-indolyl-maleinimide produced an effect comparable to the PI3K␥ inhibitor AS605240. Thus, 1 ␮M bis-indolyl-maleinimide blocked VGCC current to 51.05⫾3.05 % (n⫽9) but did not affect the ability of DAMGO to exert an acute block (Fig. 3F). Collectively these data identify PI3K␥ as an essential mediator of opioid-induced desensitization of VGCCs in small-diameter sensory neurons of DRG. In addition and in agreement with previous studies (Kelly et al., 2008), classical PKC were identified as another mediator of MOR desensitization.

DISCUSSION Involvement of PI3K␥ in the regulation of morphine peripheral analgesia has been documented in a recent study (Cunha et al., 2010). Using genetic and pharmacological approaches these authors investigated the contributions of nNOS and PI3K␥ to the acute antinociceptive effects of

C. König et al. / Neuroscience 169 (2010) 449 – 454

453

Fig. 3. Voltage-gated Ca2⫹ channels in DRG. (A) Whole-cell voltage-clamp measurements from DRG neurons according to the pulse protocol shown at the top. Currents were recorded from a wild-type cell (top) and a PI3K␥⫺/⫺ cell (bottom) before (left), about 30 s after application of 2 ␮M DAMGO (center), and upon washout (right). (B) The same protocol as in (A) but cells were preincubated for 6 h with DAMGO. (C) Analysis of the relative maximal inward current (inhibition) after DAMGO application for naive (acute) and 6 h desensitized cells (6 h des.). Data were obtained from wild-type (white), PI3K␥⫺/⫺ (grey), and transgenic PI3K␥ kinase dead (KD, black) DRG cells. The cell number analyzed per group is displayed at the bottom of each column, error bars indicate SEM values. (D) Reconstitution of PI3K␥ expression in PI3K␥⫺/⫺ DRG cells by electroporation induces robust ␮-opioid receptor desensitization by DAMGO compared to mock-transfected cells. (E) Similar to (C) but for morphine (2 ␮M) as an agonist. (F) DAMGO-induced desensitization in wild-type DRG cells without (acute) and with long-term desensitization (6 h des.) in the presence of a PI3K␥-specific inhibitor (AS605240, 1 ␮M) or the cPKC inhibitor bis-indolyl-maleinimide (Bis I, 1 ␮M). * P⬍0.05, ** P⬍0.01 in all panels.

morphine on mechanical hypernociception induced by PGE2 and other inflammatory mediators. Now, as a central finding our study discloses PI3K␥ as a mediator of tolerance development after prolonged exposure to morphine. These data expand the number of known mediators of opioid tolerance and provoke the question for underlying cellular mechanisms. The identification of PI3K␥ together with MOR in small-diameter DRG suggests a specific function of the protein in MOR desensitization in peripheral sensory neurons. These cells have been consequently selected for electrophysiological investigations.

The currently known signaling pattern of MOR is in line with the established signaling reactions of PI3K␥. Whereas MOR has been characterized as a receptor coupled to the Gi/Go subtype of heterotrimeric G proteins (Kelly et al., 2008), PI3K␥ is known as the prototype PI3K controlled by G␤␥ subunits released from this type of activated GPCR (Leopoldt et al., 1998; Hawkins et al., 2006). Direct evidence for a role of PI3K␥ as a mediator of MOR-dependent signaling was confirmed by whole-cell patch-clamp analysis of VGCC. Our electrophysiological data suggest that PI3K␥-induced signaling events enhance MOR desensitization. These data are consistent with the proposed func-

454

C. König et al. / Neuroscience 169 (2010) 449 – 454

tion of PI3K␥ in morphine-induced tolerance development in vivo. One possibility to explain the obvious involvement of PI3K␥ in morphine-induced tolerance development is that MOR-activated PI3K␥ may directly control MOR receptor desensitization. A role for PI3Ks in GPCR desensitization has been described in a number of studies (Tan et al., 2003; Patel et al., 2009). Specifically PI3K␥ reportedly mediates desensitization of the Gs-coupled ␤2-adrenergic receptor in cardiomyocytes via the direct interaction with receptor-associated protein kinase (GRK2) (Nienaber et al., 2003; Patel et al., 2009). The absence of this association may contribute to the decreased desensitization of VGCC in PI3K␥⫺/⫺ DRG induced by morphine or DAMGO (Fig. 3C–E). Another explanation for the PI3K␥-dependent enhancement of ␮-opioid desensitization is provided by the pronounced elevation of DAMGO-induced VGCC inhibition provoked by the PKC inhibitor bis-indolyl-maleinimide (Fig. 3F). Heterologous desensitization of MOR by cPKC has been proven in many models (Martini and Whistler, 2007; Kelly et al., 2008; Bailey et al., 2009). Since PI3K␥ can directly interact with and stimulate PKC␣ activity (Lehmann et al., 2009), the consecutive activation of PI3K␥ and cPKC may act as key signaling reactions of opioid-induced desensitization.

CONCLUSION In conclusion, our data suggest a specific mediator function of PI3K␥ in the desensitization of MOR- and morphineinduced tolerance development in the peripheral nervous system. These findings may open novel approaches for the therapeutic use of specific inhibitors of PI3K␥ to suppress morphine-dependent desensitization and tolerance.

REFERENCES Bailey CP, Llorente J, Gabra BH, Smith FL, Dewey WL, Kelly E, Henderson G (2009) Role of protein kinase C and ␮-opioid receptor (MOPr) desensitization in tolerance to morphine in rat locus coeruleus neurons. Eur J Neurosci 29:307–318. Bondeva T, Pirola L, Bulgarelli-Leva G, Rubio I, Wetzker R, Wymann MP (1998) Bifurcation of lipid and protein kinase signals of PI3K␥ to the protein kinases PKB and MAPK. Science 282:293–296. Bonnington JK, McNaughton PA (2003) Signalling pathways involved in the sensitisation of mouse nociceptive neurones by nerve growth factor. J Physiol 551:433– 446. Cunha TM, Roman-Campos D, Lotufo CM, Duarte HL, Souza GR, Verri WA Jr, Funez MI, Dias QM, Schivo IR, Domingues AC, Sachs D, Chiavegatto S, Teixeira MM, Hothersall JS, Cruz JS, Cunha FQ, Ferreira SH (2010) Morphine peripheral analgesia depends on activation of the PI3K␥/AKT/nNOS/NO/KATP signaling pathway. Proc Natl Acad Sci U S A 107:4442– 4447. D’Amour FE, Smith DL (1941) A method for determining loss of pain sensation. J Pharmacol Exp Ther 72:74 –79. Endres-Becker J, Heppenstall PA, Mousa SA, Labuz D, Oksche A, Schäfer M, Stein C, Zöllner C (2007) ␮-opioid receptor activation modulates transient receptor potential vanilloid 1 (TRPV1) currents in sensory neurons in a model of inflammatory pain. Mol Pharmacol 71:12–18.

Fruman DA, Bismuth G (2009) Fine tuning the immune response with PI3K. Immunol Rev 228:253–272. Hawkins PT, Anderson KE, Davidson K, Stephens LR (2006) Signalling through Class I PI3Ks in mammalian cells. Biochem Soc Trans 34:647– 662. Hirsch E, Katanaev VL, Garlanda C, Azzolino O, Pirola L, Silengo L, Sozzani S, Mantovani A, Altruda F, Wymann MP (2000) Central role for G protein-coupled phosphoinositide 3-kinase ␥ in inflammation. Science 287:1049 –1053. Kavaliers M, Hirst M (1983) Daily rhythms of analgesia in mice: effects of age and photoperiod. Brain Res 279:387–393. Kelly E, Bailey CP, Henderson G (2008) Agonist-selective mechanisms of GPCR desensitization. Br J Pharmacol 153 (Suppl 1): S379 –S388. Koch T, Höllt V (2008) Role of receptor internalization in opioid tolerance and dependence. Pharmacol Ther 117:199 –206. Koch T, Widera A, Bartzsch K, Schulz S, Brandenburg LO, Wundrack N, Beyer A, Grecksch G, Höllt V (2005) Receptor endocytosis counteracts the development of opioid tolerance. Mol Pharmacol 67:280 –287. Lehmann K, Müller JP, Schlott B, Skroblin P, Barz D, Norgauer J, Wetzker R (2009) PI3K␥ controls oxidative bursts in neutrophils via interactions with PKC␣ and p47phox. Biochem J 419:603– 610. Leopoldt D, Hanck T, Exner T, Maier U, Wetzker R, Nürnberg B (1998) G␤␥ stimulates phosphoinositide 3-kinase-␥ by direct interaction with two domains of the catalytic p110 subunit. J Biol Chem 273:7024 –7029. Martini L, Whistler JL (2007) The role of mu opioid receptor desensitization and endocytosis in morphine tolerance and dependence. Curr Opin Neurobiol 17:556 –564. Narita M, Ohnishi O, Nemoto M, Yajima Y, Suzuki T (2002) Implications of phosphoinositide 3-kinase in the ␮- and ␦-opioid receptormediated supraspinal antinociception in the mouse. Neuroscience 113:647– 652. Nienaber JJ, Tachibana H, Naga PSV, Esposito G, Wu D, Mao L, Rockman HA (2003) Inhibition of receptor-localized PI3K preserves cardiac ␤-adrenergic receptor function and ameliorates pressure overload heart failure. J Clin Invest 112:1067–1079. Patel PA, Tilley DG, Rockman HA (2009) Physiologic and cardiac roles of ␤-arrestins. J Mol Cell Cardiol 46:300 –308. Patrucco E, Notte A, Barberis L, Selvetella G, Maffei A, Brancaccio M, Marengo S, Russo G, Azzolino O, Rybalkin SD, Silengo L, Altruda F, Wetzker R, Wymann MP, Lembo G, Hirsch E (2004) PI3K␥ modulates the cardiac response to chronic pressure overload by distinct kinase-dependent and -independent effects. Cell 118: 375–387. Scherrer G, Imamachi N, Cao YQ, Contet C, Mennicken F, O’Donnell D, Kieffer BL, Basbaum AI (2009) Dissociation of the opioid receptor mechanisms that control mechanical and heat pain. Cell 137:1148 –1159. Schulz S, Mayer D, Pfeiffer M, Stumm R, Koch T, Höllt V (2004) Morphine induces terminal ␮-opioid receptor desensitization by sustained phosphorylation of serine-375. Embo J 23:3282–3289. Stein C, Lang LJ (2009) Peripheral mechanisms of opioid analgesia. Curr Opin Pharmacol 9:3– 8. Tan M, Groszer M, Tan AM, Pandya A, Liu X, Xie CW (2003) Phosphoinositide 3-kinase cascade facilitates ␮-opioid desensitization in sensory neurons by altering G-Protein-effector interactions. J Neurosci 23:10292–10301. Zhuang ZY, Xu H, Clapham DE, Ji RR (2004) Phosphatidylinositol 3-kinase activates ERK in primary sensory neurons and mediates inflammatory heat hyperalgesia through TRPV1 sensitization. J Neurosci 24:8300 – 8309.

(Accepted 28 April 2010) (Available online 6 May 2010)